English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In this paper, an unmanned aerial vehicle (UAV)-assisted wireless network isconsidered in which a battery-constrained UAV is assumed to move towardsenergy-constrained ground nodes to receive status updates about their observedprocesses. The UAV's flight trajectory and scheduling of status updates arejointly optimized with the objective of minimizing the normalized weighted sumof Age of Information (NWAoI) values for different physical processes at theUAV. The problem is first formulated as a mixed-integer program. Then, for agiven scheduling policy, a convex optimization-based solution is proposed toderive the UAV's optimal flight trajectory and time instants on updates.However, finding the optimal scheduling policy is challenging due to thecombinatorial nature of the formulated problem. Therefore, to complement theproposed convex optimization-based solution, a finite-horizon Markov decisionprocess (MDP) is used to find the optimal scheduling policy. Since the statespace of the MDP is extremely large, a novel neural combinatorial-based deepreinforcement learning (NCRL) algorithm using deep Q-network (DQN) is proposedto obtain the optimal policy. However, for large-scale scenarios with numerousnodes, the DQN architecture cannot efficiently learn the optimal schedulingpolicy anymore. Motivated by this, a long short-term memory (LSTM)-basedautoencoder is proposed to map the state space to a fixed-size vectorrepresentation in such large-scale scenarios. A lower bound on the minimumNWAoI is analytically derived which provides system design guidelines on theappropriate choice of importance weights for different nodes. The numericalresults also demonstrate that the proposed NCRL approach can significantlyimprove the achievable NWAoI per process compared to the baseline policies,such as weight-based and discretized state DQN policies.

Neural Combinatorial Deep Reinforcement Learning for Age-optimal Joint Trajectory and Scheduling Design in UAV-assisted Networks